Two 777 Engine diversions in 1 day

[RogerTangoFoxtrotIndigo]

Its being reported that UA 777 flying FRA-IAD diverted to KEF (PW powered) and a CO 777 NRT-HOU (GE powered) diverted into Midway. At a rough count that makes 4 engine diversions, all while over water in 6-8 weeks [1 GE90 into Dublin 1 GE90 into Forteleeza, Brazil]

Are we seeing something here that withe the benefit of hindsight may proove to be significant, equipment ages, and not always at the rate the manufacturer expects.

Personally I have always though that the 2 engine failure statistics are missunderstood, even if it is predicted only once in 10 million flight hours

1) It is predicted to happen sometime
2) That statistic doesnt discount 3 tomorrow and no more for 100 years!

Added to this there is something I like to refer to as Shuttle Mathematics. Back in 1980 it was predicted that the shuttle system would fail once in 10,000 launches (or something like that) after challenger that figure was reduced to 1 in less than 100, ie you would lose another one before completion of the ISS. I think the actual stats now show a 2 in 107.

What this tells us is that stats are just numbers built on assumptions and that in real life experience can sometimes proove them to be worthless especially when there is pressure for the "right result" from on high.

Getting to the point, I doubt that 2 of the above diversions were in ETOPS sectors so will not even find there way into the IFE stats, further distorting the real world reilability stats. I will also thow the Trent 800 (un)contained failure at MEL into this catagory.

Whos keeping score? how are these stats arrived at? is there any independent non regulatory input. Are the regulators to close to the Airlines?

NO, Im not a journalist otherwise the thread would have been titled 600 passengers narrowly escape death in engine fire horror.

[BRISTOLRE]

Midway should be fun for a B777!! mmmm.

[Flopster]

I may have got this wrong, however, as I understand it if you shutdown an engine for whatever reason, but do so outside the ETOPS portion of a flight, that does not count in the statistics?

If so, can someone please explain how on earth "they" came up with that idea? Technically, "you" could shut down umpteen donks outside the ETOPS areas, and yet it would have zero effect, certification wise, on "your" ability to continue operating ETOPS.

[Dewdrop]

"Midway should be fun for a B777!! mmmm"


I'm sure if you can get a fully loaded B52 in, you can manage a 777.

[MasterBates]

Apparently there was a fuel line rupture in the United 777 that landed in BIKF.
I recall this used to be somewhat a problem in the 757 (RR) few years ago, does anyone remember that?

MB.
:8

[BRISTOLRE]

Dewdrop, are we talking of the same Chicago Midway or another? Have you ever seen a B52 in there cos it has a short rwy.

[steamchicken]

In the context I think it's the one halfway across the Pacific.

[tom775257]

Midway Atoll I assume

[BoeingMEL]

Anyone fancy a wager of a few shillings? My '99 prediction that pax would get their feet wet within 5 years still stands... jeez, hope I'm proven wrong. bm

[Bagso]

wonder what the RAF and USAF will make of this...

both about to go to two engined Jet tankers .....

Bad enough with 350 pax...

but pretty horrendous if you are also towing 6 chicks (fighters)

[BigGreenPleasureMachine]

Not entirely sure of your point, Bagso, why would a failure to a twin be any worse than a failure to a quad or trijet with fighters (or anything else) refuelling? Apart from them getting offside PDQ, I can't really imagine anything more sinister with just the two powerplants.

Regards, BGPM.

[Firebird]

Being a Maint. Engineer for along time I once asked a 757 Captain did he ever lose an engine during take off ? His answer was yes, and added that using the one engine power settings actually had him climbing and speeding thereby he reduced power.
Knowing this about our new generation engines I have the utmost confidence in their power abilities.

[Old Aero Guy]

There seems to be basic mis-understanding on ETOPS statistics.
An inflight shutdown in any phase of flight becomes part of the ETOPS statistics used to determine the suitability of an engine/airframe for ETOPS operation.
Futhermore, it doesn't even need to be on an ETOPS airplane. A CFM56 shutdown on a non-ETOPS 737 becomes part of the CFM56/737 database used to determine CFM56/737 ETOPS eligibility.
Engine shutdown rates are monitored very closely by national certification agencies.
There have been instances where national authorities have removed ETOPS approval for their national carrier when the carrier's shutdown rate crept above the shutdown rate requirement.
This wasn't quite right as the shutdown rate requirement is supposed to be based on the world fleet data.
Kind of reminds you of Lake Woebegon where all of the children are supposed to be above average.

[Anti-ice]

As far as I understand firebird, the 757 had and still has the best power-to-weight ration of any airliner ever built!

Unless things have changed , this may still be true, its a v powerful bit of kit, but I think a fully laden 777 is gonna be a darn sight heavier :ooh:

The bigger they are, the harder they fall it is said, - anyone remember the (Continental?) 747 at gatwick that knocked the top of a house off after losing just one engine (25%loss) on take-off???

[RogerTangoFoxtrotIndigo]

flight sim boy: wrote

RTFI - According to the official ATSB report the emirates event at Melbourne was fully contained...

Do you know something they / we don't??

No not really, i'm just used to reading the fine print. As I understand it debris were thrown from the engine, passed under the aircraft and impacted the remaining servicabe unit. And here it is in black and white.

Failure of the RB211 Trent 892 engine as fitted to the aircraft was a result of the release of a single blade from the low-pressure compressor (fan) rotor disk. The blade release caused extensive damage to the remainder of the fan and the intake shroud, however the event was fully contained. The only escape of debris from the engine was small, low energy fragments, causing minor damage to the fuselage and the opposite engine.

I dont call that contained!

Lets think about this, on a take off run 1 engine was lost and the debris impacted the other, it didnt damage it admittedley but we are talking about the begining of the run. At or about rotation speed everything has much more energy and in a situation where you have to "go" and the other unit must be able to give you 100% there is (appears to be) a risk of debris impacting and at least passing through an engine rated and running close to 90,000 lb thrust..... Can anyone see the problem here? Whatever your TOW this is a nightmare scenario.

While on the face it seems to be a standard blade out, no drama, event looking a little closer the incident seems much more serious with disasterous potential but I dont recall any regulatory authorities working through the above and asking for assurances that the second unit could not be pinged, you have to ask why

I understood that during certification it had to proved that this could not happen but it did 2 years ago.

You also have to question the sanity of Boeing going for a single powerplant supplier on all high gross weight 777's. Not only because it p1ssed off RR customers costing them order from BA amonst others but because a proven engine related mid pacific ditching could kill the program.

[Lartington]

I recall a few years back that the aviation correspondent for the Daily Telegraph envisaged three major accidents. One was the crash of Concorde, two a midair over London and three a twin-jet going down when crossing the pond. One down two to go.

[411A]

ETOPS

Yes reasonable, for twins considering the problems involved.
Having said this, do think that anything exceeding 180 minutes is stretching it to the extreme.
Having been a 3/4 engine driver for a very long time, including trans-Pacific (where runways are not exactly expansive), suspect that if 240 minutes becomes the norm, then a wet footprint can be expected.
And also expect....the results ain't gonna be pretty.:sad:

[Anti Skid On]

Lartington

You could actually say two out of three if you include the mid air over Europe, and nearly three of three if the Air Transat A330 hadn't made land!

[tech...again]

Anti-Ice,

I thought the 747 ex LGW a few years ago lost more than one engine...? Can anyone elaborate?

TA :confused:

[cirrus01]

The CO ( exPeople Express) 747 at LGW had to have 2 engines changed before its return to USA.....

If I remember right .....One engine failed at T/O , another surged constantly......believe that a pax had a video just after T/O of the No. 4 with plenty of flames from the jet pipe and the ground alarmingly close.....

Max weight, Max X-wind, ........also seem to recall that flight was to Maimi, but was alway flight planned to Washington, as it could always make that distance, and then if the gods were favourable it would refile during the flight to Maimi.

Anyone confirm this ?
Any practices like this still continue ??

[Mode7]

Old Aero Guy. Where did you get this data from? I was under the impression that if ain't in the ETOPS sector it doesn't count in the ETOPS stats but in others.

[lomapaseo]

by RTIF

I dont call that contained!

Lets think about this, on a take off run 1 engine was lost and the debris impacted the other, it didnt damage it admittedley but we are talking about the begining of the run. At or about rotation speed everything has much more energy and in a situation where you have to "go" and the other unit must be able to give you 100% there is (appears to be) a risk of debris impacting and at least passing through an engine rated and running close to 90,000 lb thrust..... Can anyone see the problem here? Whatever your TOW this is a nightmare scenario.

While on the face it seems to be a standard blade out, no drama, event looking a little closer the incident seems much more serious with disasterous potential but I dont recall any regulatory authorities working through the above and asking for assurances that the second unit could not be pinged, you have to ask why

I understood that during certification it had to proved that this could not happen but it did 2 years ago.


Lets try to keep to facts in evidence and our dreams and imaginations out of safety assessments.

First off the regulations specifiy a level of containmnet and not complete containment of every little bit each and every time. Therefoe a further level of robustness is expected at the aircraft design level most of the time. Thus we have the relatively good history, including the Trent event in OZ land.

In the structural part of the safety equations we are doing quite well compared to all other causes (weather, operation, human mistakes etc.)

The arguments about ETOPS are now long in tooth and while everybody expects a splash someday, the odds are it will be a 4 engine machine if its engine related and not a twin.

[RogerTangoFoxtrotIndigo]

Lets try to keep to facts in evidence and our dreams and imaginations out of safety assessments.

~~~~Snip~~~~

The arguments about ETOPS are now long in tooth and while everybody expects a splash someday, the odds are it will be a 4 engine machine if its engine related and not a twin.

Ok, apart from the fact I find the begining and the end of your post hard to reconcile, Its not my imagination its in the report. Debries did impact the other engine, now if thats the standard for a "contained" engine failure I guess so be it, but as you are aware there are very few fatal accidents that were not preceeded by less serious incidents that should have served as warnings.

The question I was asking at the head of the topic was should we be looking closer at these failures. As I understand it all of these recent failures have been due to different causes and that from a statistical and materials point of view gives more cause for concern not less.

[lomapaseo]

by RTFI

The question I was asking at the head of the topic was should we be looking closer at these failures. As I understand it all of these recent failures have been due to different causes and that from a statistical and materials point of view gives more cause for concern not less.

OK

Fair question.

I would be more concerned about failures for the same cause, as that indicates a growing specific problem whose specific soulution his not working. This is under the presumption that all known specfic problems have been addressed with a specific corrective action.

OTOH, multiple scattered problems are symptomatic of general wear and tear not addressed by maintenance. And as such these problems are expected to be random and best tracked by statistics up to a critical level.

How they are to be addressed is another subject entirlely

[Divergent Phugoid!]

Anti Ice....

Its onle been a matter of weeks but you forget about Concorde.
That surely has to have the highest power to weight ratio especially with burners on!!;)

[DJohnsen]

Mode7: Old Aero Guy. Where did you get this data from?

"Old Aero Guy" is absolutely correct and you can find this in FAA AC 120-42A (http://www.airweb.faa.gov/Regulatory_and_Guidance_Library/rgAdvisoryCircular.nsf/0/2638EAF8B89680A8862569BA00751C8C?OpenDocument)

The IFSD rate for a specific aircraft/engine combination is an "industry" rate as well as an "individual operator" rate... and any IFSD… see definition… will count as an event… even if you “inadvertently” shut the engine down. ETOPS is not just an aircraft/engine reliability program… but the Airline’s operational philosophy as well. Exceeding the IFSD “alert" level… say 0.02 for 180 minutes ETOPS… does not necessarily mean that you loose your ETOPS approval either, but you have to come up with a good “game plan” and convince the “authorities” you know what you are doing.

All you guys who “question” ETOPS operation should take time and read through this AC and try to understand all that is involved. I doubt any carrier will launch off across the North Pole without first study “every” available aspect of this operation… and if they don’t… well... then they are likely to “jeopardize” safety in many other areas as well… :ok:

Thought and comments…?

Dag

[B767300ER]

RTFI---

I assume you meant CO 777 was flying NRT-IAH, not Houston-Hobby airport (HOU).

And yes, Midway Island in the Pacific was where they went. Not too bad for a 777.

UAL still has the record---a 193-minute ETOPS diversion AKL-LAX into Kona,Hawaii last year. THAT was long!

[mgc]

RTFI

you are probably right about engine failure statistics being misunderstood. There is a consulting engineering company based in Guildford who specialise in risk analysis. They spend a lot of time explaining to people that a 1 in 10 000 000 hour event has nothing to do with when an event will occurr or on the likelyhood of it happening tomorrow. In fact they can do some very interresting risk calculations that build in the real life factors that can give some very different and vey un-nerving answers.

As a general rule of low risk of occurance with high consequences if it does requires a very high 1in x figure to be quoted. I am not sure 1 in 10 000 000 as you quote is particularly reassuring or that it actually means anything!:sad:

[RogerTangoFoxtrotIndigo]

Been doing a bit more research into this, the actual permitted IFSD rate to retain 180min ETOPS is .02/1000 flight hours (1 in 50,000 flight hours) while crawling the web I came across this article written by Joseph P Marksteiner who is / was manager of investigations and accident prevention at GE

Also found an article on GE's website dated 22/10/03 stating that that the GE90 had passed 4,000,000 flight hours on the 777. Ill let someone else do the maths, as its 03:35 here. I did them and came up with some scary figures for permitted IFSD's without losing ETOPS status, would like someone else to try and work through them.

Permitted shutdown rate .02/1000 hours - 80,000 engine hours per month - 126 ships in service - 4 million flight hours total - 8 years service.

ARE AIRCRAFT MAINTENANCE PRACTICES TOO RIGOUROUS?
May some current maintenance practices and philosophies be causing more problems than they are preventing? Studies have suggested that nearly a third of in-flight shutdowns (IFSD) are caused by errors brought about by maintenance activities.

Joseph P Marksteiner, GE Aircraft Engines

As aircraft and systems age, the need for preventative and corrective maintenance activity usually increases. "Airlines, aircraft manufacturers and government regulators jointly work out detailed scheduled maintenance programs designed to avoid and catch problems before they become serious enough to jeopardise an aircraft's ability to fly safely… At each step in the process mechanics probe deeper and deeper into an aircraft taking apart more and more components for closer inspection."

This increased level of maintenance activity, while reducing risks due to equipment failure, dramatically increases the risk associated with maintenance activity itself.

Sources
ARE AIRCRAFT MAINTENANCE PRACTICES TOO RIGOUROUS? (http://www.airport-technology.net/informer/management/aircraft/)

GE engines surpass 4 million flight hours powering 777 (http://www.geae.com/aboutgeae/presscenter/ge90/ge90_20031022.html)

[mgc]

one of the things the company ,I refered to in my last post, has been known to do is factor in the induced failure rate caused by human interferance, ie maintenance! This can give some really scary results.:uhoh:

[oceancrosser]

Flying the Boeing twin twins (757/767) my company has had 2 engine shut-downs in the last 3 years. I was flying in one of those cases. Turns out Maintenance installation errors and sloppy work (and more sloppy quality control) was to blame in both cases. Perhaps more focus is needed on hangar work?

[Old Aero Guy]

RTFI

It isn't a Cert. requirement that an uncontained engine failure will not cause power loss on other engines.
The standard is that an uncontained engine failure (which is already a very low probability event) only has a 1 of 20 chance of causing loss of the airplane.
My understanding of history is that this requirement evolved out the the Concorde cert. since the proximity of engine pairs made the airplane the equivalent of a twin in interms of collateral engine damage, although it was operating to the rules of a quad.

The eye-openning information you may have computed is that in 4,000,000 engine hours, 80 IFSD's would have been allowed without violating 180 min. ETOPS reliability requirements.
In fact, although .02 per 1000 engine hrs. are the standard for 180 min. ETOPS, actual fleet performance for ETOPS twins is about .005 per 1000, or about four times better than necessary.
Therefore even two IFSD's in a month don't change the statistics very much.

lomapaseo puts things in the correct context when he says the real concern would be if there were a common cause for a string of IFSD's.

PS to DJohnsen, thanks for the backup reply.

[bcp7]

CO operates daily 777 service between IAH and NRT . For the incident mentioned, the aircraft had a loss of oil quantity and oil pressure on one engine. The crew did an in-flight shutdown iaw their procedures, and diverted to their ETOPS alternate, PMDY, Henderson Field, Sand Island, Midway Atoll, in the Pacific Ocean.

Midway Atoll is roughly 1150 Nautical miles from Honolulu, and the airport has limited facilities for passenger handling.

[Buster Hyman]

Must admit, I'd never read the ATSB report, but the contained failure bit made me laugh. Having seen the EK aircraft after the event, if that is defined as uncontained, then gawd elp us!

At the time, I'd spoken to some CX engineers about the incident & they were quite concerned as to the amount of damage to the opposite engine. Is there a case for "Concorde syndrome" here?:rolleyes:

[RogerTangoFoxtrotIndigo]

OldAeroGuy

Thats one of the eye opening stats along with if you spread these faiuures out over a 126 ship fleet nearly 2/3rds of them (63%) could have suffered an IFSD without losing ETOPS certification and that with 80,000 hours per month we should expect up to 1.6 shutdowns per month. Which means that the 3 GE90 IFSD over the last 6-8 weeks are nothing to be concerned about!(?) At least according to the rules.

In fact, although .02 per 1000 engine hrs. are the standard for 180 min. ETOPS, actual fleet performance for ETOPS twins is about .005 per 1000, or about four times better than necessary.

I will direct you to Federal Register / Vol. 68, No. 114 / Friday, June 13, 2003 ... (http://a257.g.akamaitech.net/7/257/2422/14mar20010800/edocket.access.gpo.gov/2003/pdf/03-14992.pdf) which, states FAA computed shutdown rates for 1997/98 for all three certified powerplants implies that the GE at least was mighty close to breaching even that 0.02 /1000 rate. It is a facinating document for many reasons I recomend it. FAA quoted in context..

The Boeing Model 777–200 series airplane powered by Pratt & Whitney PW4077 engines was approved for ETOPS on May 30, 1995 and entered service in June 1995. By all accounts, it was a very successful new model introduction. This was followed by ETOPS approval of the 777–200 powered by General Electric GE90–77B and Rolls-Royce RB211–Trent 877–17 engines in October 1996. The inflight shutdown (IFSD) rate for all three engine types was zero for at least the first year in service. The Pratt & Whitney PW4000 reached a peak 12-month rolling average engine IFSD rate of .018/1000 hours in October 1996. The General Electric GE90 reached a peak of .021 for one month in July 1998 and the Rolls-Royce Trent reached a peak of .016 in December 1997. Although the inflight shutdown rates stayed within the allowable .02/1000 hour standard for 180-minute ETOPS, significant design problems were discovered on each engine type after ETOPS approval. During the first two years after ETOPS approval of each engine type on the Model 777 series airplanes, the FAA was concerned that the design problems being discovered may have indicated a failure of the early ETOPS process to identify those failure modes before they occurred in service. Some failure modes had the potential to result in inflight shutdowns had they occurred under different circumstances or had they not been detected during maintainence for unassociated reasons..

Thanks to all who are taking the time to contribute to this thread, 3 pages and no flames, remarkable!

[Dream Land]

Continental Airlines Boeing 777 emergency landing on Midway



By ROSANNA RUIZ - Copyright 2004 Houston Chronicle

Oil spewing from one of its engines, the limping airplane managed to
find
the tiny airfield at Midway Atoll.

From her window seat aboard the twin-engine Continental Boeing 777,
college
student Dawn Smith, 23, had a view of the ailing engine as she prepared
for
the landing on Midway's Sand Island airfield early Tuesday. The landing
made
for uncomfortable moments.

"The short runway was not built for a large aircraft," Smith, who
looked
flushed with her mussy hair, said Wednesday after the plane finally
made it
to Houston a day later than expected.

The 11-hour, 40-minute flight from Japan's Narita Airport to Houston's
Bush
Intercontinental Airport was due to arrive at 3:30 p.m. Tuesday.

Continental Flight 6 had 279 passengers and 15 crew members. A
Continental
spokeswoman previously said there were 235 passengers.

No injuries were reported.

The plane made its emergency landing at 3:10 a.m. Tuesday, Midway time.
The
atoll is about 2,200 miles east of Japan and near the northwestern end
of
the Hawaiian Islands archipelago.

After repairs were made to the engine's starter, the plane left the
atoll at
2:30 a.m. Wednesday.

Smith, a Sam Houston State University history major, had studied the
island's history as a key site of a U.S. naval victory during World War
II.
She took the unexpected 24-hour layover in stride.

Still, the food rations and uncomfortable sleeping arrangements inside
a
musty theater did not make for the most stellar accommodations.

"It'll be a long time before I get on another airplane," Smith said
upon her
arrival at Bush on Wednesday. She had spent the holidays with her
sister,
who is serving aboard the USS Kitty Hawk.

Smith and the others waited a few hours aboard the plane before a
ladder was
found to get them off the aircraft. The passengers were given
sandwiches,
candy bars and beverages and were allowed to tour Midway Atoll National
Wildlife Refuge, which is managed by the U.S. Fish and Wildlife
Service.

The pristine beaches, lush tropical foliage and millions of birds, such
as
the Laysan albatrosses or "gooney birds," Hawaiian monk seals and other
animals led Smith to conclude that "God is so wonderful."

"It was something to see," said another passenger, Allen McMahon, a
61-year-old Montgomery resident, of the onslaught of almost 300 people
to an
island that typically is populated by about 30 people.

McMahon's wife, Judy, 57, began to cry as she embraced her husband, who
appeared a little travel-weary. "I'm just glad to have him home," she
said.

Julie King, a spokeswoman for Houston-based Continental, said customer
service officials had met all passengers when they arrived at Bush.
Some
passengers upset with the inconvenience were offered travel
certificate.



Dream Land
:ooh:

[Jim Morehead]

Although I have never been to Midway, I've flown over it.

Is it possible for 4 engine airplanes to go there for 24 hours at the beach and the gooney birds??? What if the aft coffee maker fails? Would that qualify?

[JamesT73J]

"The short runway was not built for a large aircraft."

I thought it was quite a generous size?

[RiverCity]

The pristine beaches, lush tropical foliage and millions of birds, such as the Laysan albatrosses or "gooney birds," Hawaiian monk seals and other animals led Smith to conclude that "God is so wonderful."

Up to that point, I was going to send a donation to her favorite 'bad hair' charity. But then I thought, "You know, she could have been stuck here in Wilkes-Barre PA with naught but worked-out coal mines, pigeons and a boarded-up downtown. I bet her father called her 'Princess' just once too often."

[EFP058]

"The short runway was not built for a large aircraft."

I thought it was quite a generous size?

Don´t you dare to doubt the expertise of a 23 year old history student when it comes to emergency landings. Keep in mind that she did have a window seat after all.

[Paul Wilson]

Acording to http://www.airnav.com/airport/PMDY runway length is 7904ft, of course we can't judge the student too harshly, airport/runway seem to be interchangable to the media and general public, and it cannot be denied that airport is small, with airbridges and redcaps in short supply;)

Just seen the fuel price at $2.75, that must have been painful, plus they had to fly in a mechanic, parts, and 300 boxed lunches. Ouch

[hobie]

thank you Dawn Smith for telling us your story ........

http://www.history.navy.mil/photos/images/g450000/g451086.jpg

hobie ......

[av8boy]

Landed there no-flaps in a C141 a couple of decades ago. I think I survived... Lemme check... Then again, I didn't major in history so I probably didn't understand the danger I was in.

:O

Dave

[gofer]

Boeing Specs indicate

Max Landing Weight, Wet runway, Sea Level - Min is a shade under 7300Ft for a 777-300ER of course dry and empty its a shade under 5000Ft

But then that is History... although at 2.75 you don't really have to dump too much...... O

[Desert Nomad]

Possibly a little off subject and certainly not 100% of the facts but last July flying DXB LHR on a 777 rolling down the runway when starboard engine goes bang and we slew all over the shop and finally come to rest. No smoke or fire but a fairly worrying incident none the less and plenty PAX slightly more concerned. Airline excellent and had us back up within the hour (new aircraft).

On return I checked up on the report and can't remember the exact details but seem to remember approx 50 knots short of take off speed.

Couple of questions. Is this close to take off speed and what is likely outcome should that have happened just when wheels leaving ground (I know much depends on reactions etc)?

Cheers

DM

[lomapaseo]

As the wheels leave the ground expect some roll as well, all easily correctable at that slightly higher speed.

The really good news is that you already have the nose pitch uo and you will just scoot right out of there.

[Gertrude the Wombat]

Added to this there is something I like to refer to as Shuttle Mathematics. Back in 1980 it was predicted that the shuttle system would fail once in 10,000 launches (or something like that) after challenger that figure was reduced to 1 in less than 100, ie you would lose another one before completion of the ISS. I think the actual stats now show a 2 in 107.

What this tells us is that stats are just numbers built on assumptions and that in real life experience can sometimes proove them to be worthless especially when there is pressure for the "right result" from on high.'S easy. You start with the target figure, eg the marketroids or politicians tell you that you need to be able to quote a system failure rate of 1 in 10,000.

There's mathematics that lets you calculate the system failure rate from the MTBF (and other more exotic parameters) of the components and so on. However, in these cases, you already know the system failure rate so you run the maths backwards to get the MTBF etc of the components.

Then you write these figures into the contracts with the suppliers, eg someone selling you a nut and bolt has to sign up to a MTBF of two and a half trillion years. By this stage the required numbers are clearly ludicrous, so the customer can't possibly really mean it can they, so the supplier just shrugs and signs. At this point the customer has a paper trail "proving" that the required system failure rate has been delivered.

Yeah, right. So that's how it works when specifying components for, say, a CD player, OK, so it breaks, what the hell, you bin it and buy a new one. Couldn't happen for safety critical stuff, now, could it??

[Lu Zuckerman]

To: Gertrude The Wombat

What this tells us is that stats are just numbers built on assumptions and that in real life experience can sometimes proove them to be worthless especially when there is pressure for the "right result" from on high.

Are you a Reliability and /or Safety engineer? I have been a Reliability, Maintainability and Systems Safety Engineering consultant since 1968 and the above statement is 110% correct.

:E :E